Gentrification and Sustainability

Our research is grounded in the sustainability imperatives outlined by the United Nations Sustainable Development Goals (SDGs), particularly Goals 1 (No Poverty), 3 (Good Health and Well-Being), 10 (Reduced Inequalities), and 11 (Sustainable Cities and Communities).⁴. These goals provide a framework for understanding and addressing the multifaceted challenges posed by gentrification in Brooklyn. For instance, as gentrification often exacerbates economic disparities and contributes to the displacement of marginalized communities, our investigation into its causes and impacts directly aligns with the SDG 1 goal of poverty reduction and SDG 10 goal of reducing inequalities. Additionally, the SDG 11 goal of sustainable cities and communities underscores the importance of urban development that promotes inclusivity, resilience, and environmental sustainability, concepts that are central to our inquiry.

Furthermore, our research is guided by the principles of a Just Transition, as advocated by the Climate Justice Alliance.⁵ These principles emphasize the equitable redistribution of resources and power, aiming to address and transform existing social inequities based on race, class, gender, and other forms of oppression. By examining the historical, economic, and policy factors driving gentrification in Brooklyn, we seek to uncover the systemic injustices perpetuated by current urban development paradigms and explore pathways towards more just and sustainable outcomes. In doing so, our research not only contributes to the broader discourse on urban sustainability but also advances efforts to create communities that are equitable, inclusive, and environmentally resilient.

Zoning Policies

Figure 7: Barchart of the percent that each community district residential area was classified “Declining” or “Hazardous”.⁹

Redlining was a discriminatory practice employed by the federal government and banks in the 1930s, during the Great Depression, aimed at determining loan guarantees based on neighborhood characteristics and residents’ race. The Home Owners Loan Corporation (HOLC) created “residential security maps” that classified neighborhoods into categories such as “Best,” “Still Desirable,” “Definitely Declining,” and “Hazardous,” with racialized descriptions like “infiltration of Negroes” used to designate areas as lower value. This practice systematically denied people of color access to mortgage loans, perpetuating racial segregation and depriving them of resources and opportunities available to white residents. Redlining concentrated people of color in underinvested neighborhoods while providing more resources and opportunities to white residents, reinforcing systemic racism and perpetuating economic disparities¹⁰. Many areas of Brooklyn were redlined, and Figure 7 shows the percentage each community district was classified as “Declining” and “Hazardous.” This plot was constructed using redlining data from NYC’s Environment and Health Data Portal, pivoted from wide to long, and cleaned for better usability.The areas that were identified to be the most “Hazardous” were Bedford-Stuyvesant, Greenpoint and Williamsburg, Brownsville, Fort Greene and Brooklyn Heights, and Park Slope and Carroll Gardens. Meanwhile Borough Park, Bensonhurst, Sunset Park, Bushwick, Crown Heights and Prospect Heights.

Tax Incentives

One factor that may have heavily spurred gentrification in NYC, and particularly Brooklyn, is tax exemption programs created by the government that provide benefits to developers creating new residential units. In order to stimulate the housing market and incentivize developers to create housing options suitable for lower income citizens, several tax exemption policies were created that would partially exempt developers from property taxes on properties that were enrolled in the program. One common tax exemption program is 421a; in order to qualify, the property must have a certain percentage (roughly 25% or higher) of units set aside for affordable housing/low-middle income households and set them to prices that are affordable for families earning up to a certain percentage of AMI (Area Median Income).

Figure 8: Bar Graph depicting Housing Units qualified for varying tax exemption programs per borough

Many criticize the 421a tax exemption program, claiming that despite the program’s intent to increase affordability of rent, it may have the opposite effect. This may be due to some developers taking advantage of the requirement to set housing prices based on the area’s AMI and raise the rent to cater to those households who earn 120-130 percent of AMI.

Figure 9: Line Graph depicting Median Gross Rent by Neighborhood in Brooklyn

A majority of Brooklyn neighborhoods have a similar, steady increase in median rent over the years, such as Flatlands, Boroughs Park, Sheepshead Bay etc. Williamsburg and Brooklyn Heights, on the other hand, have had a steeper shift in rent; both had been on par with the median rents of other Brooklyn neighborhoods in 2006, but ultimately are now amongst the top 3 most expensive neighborhoods.

Figure 10: Bar Graph depicting difference between low income units and 421a tax units per borough built between 2010-2021

Figure 10 was created by joining two datasets from NYU’s Furman Center and summing over the total number of each unit type for each borough. The total number of low income units per borough was overlaid over the total number of 421a qualified units per borough. The Furman Center defines low income units in this dataset as any unit “targeted to households that earn less than/equal to 80% AMI.” This would include units that are also 421a qualified.

Comparing each borough’s ratio of 421a-qualified units and low income units, we see that the borough with the largest disparity between 421a units and low-income is clearly Brooklyn, which has an incredibly large number of 421a built units in comparison to the number of low income ones. Brooklyn also has the largest population of the boroughs. We cannot necessarily draw any conclusions from this graph that the 421a exemption program is causing gentrification in Brooklyn, especially since this dataset only shows the total number of units built between 2010 and 2021, and does not include more detail on how these ratios have changed. However, it is interesting to compare Brooklyn’s ratio to other boroughs since the gap is much larger.

Figure 11: Linear Model of Number of 421a Units in Brooklyn’s Community Districts and the Median Rent Increase since 2006 to 2021

Figure 11 was created by joining together two separate datasets from the Furman Center’s datahub CoreData. The median rent data originally contained the values of rent per community district across the years 2006-2021, so in order to calculate the rent increase across this time period, the difference was taken for the rent value of 2021 and 2006 for each community district. The total number of 421a units built was already for the year 2021. After fitting a linear model to the two variables, we can see that there is a positive correlation with relatively small residuals. This indicates that the correlation between a community district having a large number of 421a units and the median rent increasing is strong. With the data, however, we cannot make any conclusions about causation. However, it is also important to note that there are only 18 data points, one for each community district. Ideally, we would be able to have more data points to strengthen our conclusion.

Displacement of Public Housing

Figure 12: Barchart showing the number of public housing developments in each of the five boroughs of NYC¹¹

As shown by Figure 12, Brooklyn offers just over 60 public housing developments, despite having the largest population of the five boroughs at 2,600,000+ people¹². The Bronx has a much lower population of 1.5 million, and Manhattan has 1.7 million people. Brooklyn also has the second largest land area of the boroughs at 71 square miles, following Queens at 109 square miles¹³. Considering the abundance of people that live in Brooklyn and the large amount of land available, this borough should surpass Manhattan and the Bronx in public housing developments. This plot was made using data from New York City Housing Authority’s Public Housing Development database. This data was cleaned and grouped to organize the visualization by borough.

As shown in Figure 13, Brooklyn severely lacks public housing developments, especially in the bottom half and center of the borough. Some public housing is available in Coney Island, Brownsville, Canarsie, Bedford-Stuyvesant, and Downtown Brooklyn. This map was also made using data from New York City Housing Authority’s Public Housing Development database in combination with leaflet.

Figure 14: New construction of government owned housing projects by borough. Affordable housing is HPD (until 2015), NYCHA, and NYCHA/HHC. Note lack of new projects after 2015 in Brooklyn compared with other Boroughs.

Figure 14 shows the construction of new government housing. HPD (Housing Preservation and Development) and NYCHA (New York City Housing Authority) represent affordable housing projects. Though 2015 saw a great deal of new housing being built in Brooklyn, since then very little to no development has occurred, while other boroughs have seen new developments.

Real Estate Speculation

Figure 15: Percent ownership of private for-profit housing in NYC. Note the decreasing share of individually owned properties in Brooklyn.

Figure 15 shows that new buildings that are individually owned have been going down over the years as a percentage of the total new houses. Compared to other boroughs (Bronx and Queens) this trend is unique to Brooklyn. This points to the fact that in Brooklyn, most new houses are owned by corporations and partnerships, a symptom of real estate speculation that can drive out families from their properties.

Real estate speculation involves the practice of treating property or housing as a financial investment and not as a purchase of a place to live. While speculation can take the form of buying land well in advance of anticipated development, we will focus on a narrower slice of this world: house flipping. House flippers purchase a house, renovate or retrofit it to update its design, and resell it for profit, hopefully making more than enough money to offset the upgrades.

A common policy practice used by local governments to reduce the influence of flipping on the housing market is an elevated tax on properties sold under a minimum amount of time. It would follow from this that a key indicator of flipping, then, could be a large number of properties sold within a shorter range of time. Flipping generally occurs within one year of purchase, but larger projects may extend beyond this range.

Figure 16: Property hold times in the Bedford Stuyvesant neighborhood of Brooklyn for properties that were sold at least twice between January 1, 2013 and December 31, 2023.¹⁵

When comparing the differences between consecutive property sale dates in Bedford Stuyvesant, the short term holding phenomena becomes clear. The trend is clear that properties in Bed-Stuy have been resold relatively quickly. Greenpoint, however, does not display a similar skew toward lower hold times, despite being marked as a gentrifying region. Flipping, therefore, is not always a symptom or cause of gentrification, and other factors may be at play.

Another way to evaluate the presence of flipping is to look at value appreciation on a short time scale. This will reveal whether flippers are successfully inflating the value of their properties faster than they would appreciate normally. A flatter correlation could suggest that flippers are making a similar return on their investment to homeowners who hold onto them longer, and a positive correlation would suggest that flippers gain no outsize value above simple appreciation.

Figure 17: Appreciation data by how long the property was held. Vertical axes are log scale to better illustrate distribution. Data was filtered for hold times greater than 60 days. [ibid]

Neither of the plots in Figure 17 clearly reveals a very strong correlation between the housing price and the time held. Notably, both plots represent an elevated initial appreciation, followed by a dip, that then morphs into a more linear growth. This supports the narrative that gentrification is occurring because flippers are experiencing greater appreciation than those who hold their houses for longer.

Transportation Infrastructure

Upon research of transportation and gentrification, it is found that the literature is split on the relation between transportation and gentrification.An article from Bloomberg studied whether transit corridors contribute to the displacement of low-income households.¹⁶. Although there is an initial correlation between high incomes and access to subway stops, a more detailed analysis showed that transit access is not the primary factor driving changes in neighborhood income levels between 2000 and 2010. Instead, other factors such as population growth, racial diversity, education levels, and proximity to central business districts play more significant roles. The study highlights the importance of transit for less advantaged groups and suggests that a lack of transit access can hinder economic opportunities for these groups.

Another study from University of Texas, Austin¹⁷ examined the relationship between gentrification and public transit service in the New York–Northern New Jersey–Long Island area. It identified higher rates of transit deserts (areas with limited transit service) in economically disadvantaged neighborhoods and highlighted the significant impact of transit services on gentrification progress. However, the study also noted that transit services have insignificant effects on super-gentrification. Policymakers are urged to allocate public transit budgets carefully to avoid displacement while meeting the transit needs of different socioeconomic groups.

To corroborate information found in these articles, Figures 18 and 19 were created to explore the relationship between subway station count per neighborhood in Brooklyn and different factors that indicate the occurrence of gentrification.

Fig 18. Map of subway count by neighborhood along with three factors that are often a sign of gentrification. Each map is overlaid with subway transit points.

Figure 18 was created by combining data from the Census American Community Survey¹⁸, Department of Investigation¹⁹, and MTA Subway Stations²⁰. Data was combined and summarized to get the final counts shown in Figure 18. By comparing the shaded areas of the various maps in Figure 18, a correlation is exposed between subway count and mean rent. As the number of subway stations increase in a neighborhood, so does the mean rent.

Fig 19. Map of subway count by neighborhood along with demographic data. Each map is overlaid with subway transit points.

By comparing subway station count and the various demographics, Figure 19 shows that the neighborhoods with the highest densities of Black people have the lowest amount of subway stations.

Overall, although the graphs presented in Figure 18 and Figure 19 show some correlation between transportation and gentrification, there are some limitations to the data. These limitation include the fact that different data sets are taken from different years based on available data, and the fact that it just accounts for how many subway stations there are per neighborhood rather than proximity to subway stations.

Future Work

If we had more time with this project we might expand our research to look at all boroughs in New York City and look at a more expanded timeline of data to fully depict the story of gentrification in New York City.